lang.j.programming (date)

December 31, 2006

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: Oleg Kobchenko <olegykj@xxxxxxxxx>

In-reply-to: <504998.90347.qm@xxxxxxxxxxxxxxxxxxxxxxx>
References: <504998.90347.qm@xxxxxxxxxxxxxxxxxxxxxxx>

I am not sure if qscreen should be among wd
commands at all.

"qscreen" should be called qdevice (or qcontext),
since it's applicable to selected output media
like printer. Thus it is more appropriate
for gl2 commands, where destination can be a printer.

Here's usecases from J library

g:\math\j601\system\classes\view\viewmat.ijs (1)
  max=. 2 3 { 0 ". wd 'qscreen'

g:\math\j601\system\examples\games\solitaire\main.ijs (1)
xy=. 2 3{".wd'qscreen'

g:\math\j601\system\examples\games\solitaire\options.ijs (1)
xy=. 2 3{".wd'qscreen'

g:\math\j601\system\extras\util\debugs.ijs (1)
5 { 0 ". wd 'qscreen'

All but last should use qm. Last is probably
a hack left from the old 72/96 dpi hacking days,
fixed in j601; thus should not be used.

Removing qscreen from wd and make an appropriate
command in gl2 would eliminate this confusion.

Not all things are needs like millimeters.
Note: in Java the only useful thing it reports
is DPI, which again makes more sense in gl2
in the context of current device like screen, 
paper etc.


--- Oleg Kobchenko <olegykj@xxxxxxxxx> wrote:

> Esp. that you are comparing output of
> qscreen with definition of qm.
> 
> --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> 
> > I think I'm supposed to understand this stuff, but
> > I just don't know what the words mean...    in
> > 
> > C:\J\system\extras\help\user\wd_commands.htm
> > 
> > I see
> > 
> > qm ; return system metrics:
> >         screen width, screen height,
> >         x logical unit, y logical unit,
> >         cxborder, cyborder,
> >         cxfixedframe, cyfixedframe,
> >         cxframe, cyframe,
> >         cycaption, cymenu,
> >         desktop x, desktop y,
> >         desktop w, desktop h
> >         More elements may be added.
> > 
> > 
> > I just don't know what this refers to.  For example,
> > 'x logical unit'.  I don't know if this is describing
> > something in logical units (in which case my question
> > ins, What's a logical unit?) or telling me what a logical
> > unit is (in which case I have the additional question,
> > What is it that is being measured in logical units?).
> > 
> > On my system I get
> > 
> >    wd 'qscreen'
> > 362 290 1280 1024 96 96 32 1 _1 36 36 51 
> > 
> > 12 values, which doesn't match the 16 things listed above.
> > 
> > I have similar basic questions about wd 'qscreen'.
> > 
> > 
> > I just have the feeling this stuff must be described somewhere.
> > 
> > Henry Rich
> >  
> > 
> > > -----Original Message-----
> > > From: programming-bounces@xxxxxxxxxxxxx 
> > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Oleg Kobchenko
> > > Sent: Sunday, December 31, 2006 3:25 PM
> > > To: Programming forum
> > > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > > 
> > > Yes to both, it's explained in WD Commands page.
> > > 
> > > The things they relate to is the basic
> > > UI stuff, like the size of the window title bar etc.
> > > 
> > > --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> > > 
> > > > Wait... is it the reply to 'qm'?  What do all those things mean?
> > > > And what does the reply to 'qscreen' mean?
> > > > 
> > > > I'm guessing there is some Windows information page that has all
> > > > these definitions, and I hope someone can point me to it.
> > > > 
> > > > Henry Rich 
> > > > 
> > > > > -----Original Message-----
> > > > > From: programming-bounces@xxxxxxxxxxxxx 
> > > > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Henry Rich
> > > > > Sent: Sunday, December 31, 2006 2:37 PM
> > > > > To: 'Programming forum'
> > > > > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > > > > 
> > > > >  
> > > > > 
> > > > > > See screen metrics.
> > > > > 
> > > > > That sounds useful.  Where is it?
> > > > > 
> > > > > Henry Rich
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <791001.90770.qm@xxxxxxxxxxxxxxxxxxxxxxx>
References: <791001.90770.qm@xxxxxxxxxxxxxxxxxxxxxxx>

Thanks... and wd 'pmove' and wd 'pmovex' will work
for me.

Henry Rich

> -----Original Message-----
> From: programming-bounces@xxxxxxxxxxxxx 
> [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Oleg Kobchenko
> Sent: Sunday, December 31, 2006 3:59 PM
> To: Programming forum
> Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> 
> You don't need to know.
> 
>    wd'qm'
> 1280 1024 8 16 1 1 3 3 4 4 22 20 0 0 1280 994
> 
>    wd'pmovex 0 0 ',": _2{.". wd'qm'
> 
> will resize it to full desktop, which is
> smaller than screen.
> 
> I get it not from knowledge, but from observation.
> 
> 
> --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> 
> > I think I'm supposed to understand this stuff, but
> > I just don't know what the words mean...    in
> > 
> > C:\J\system\extras\help\user\wd_commands.htm
> > 
> > I see
> > 
> > qm ; return system metrics:
> >         screen width, screen height,
> >         x logical unit, y logical unit,
> >         cxborder, cyborder,
> >         cxfixedframe, cyfixedframe,
> >         cxframe, cyframe,
> >         cycaption, cymenu,
> >         desktop x, desktop y,
> >         desktop w, desktop h
> >         More elements may be added.
> > 
> > 
> > I just don't know what this refers to.  For example,
> > 'x logical unit'.  I don't know if this is describing
> > something in logical units (in which case my question
> > ins, What's a logical unit?) or telling me what a logical
> > unit is (in which case I have the additional question,
> > What is it that is being measured in logical units?).
> > 
> > On my system I get
> > 
> >    wd 'qscreen'
> > 362 290 1280 1024 96 96 32 1 _1 36 36 51 
> > 
> > 12 values, which doesn't match the 16 things listed above.
> > 
> > I have similar basic questions about wd 'qscreen'.
> > 
> > 
> > I just have the feeling this stuff must be described somewhere.
> > 
> > Henry Rich
> >  
> > 
> > > -----Original Message-----
> > > From: programming-bounces@xxxxxxxxxxxxx 
> > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of 
> Oleg Kobchenko
> > > Sent: Sunday, December 31, 2006 3:25 PM
> > > To: Programming forum
> > > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > > 
> > > Yes to both, it's explained in WD Commands page.
> > > 
> > > The things they relate to is the basic
> > > UI stuff, like the size of the window title bar etc.
> > > 
> > > --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> > > 
> > > > Wait... is it the reply to 'qm'?  What do all those things mean?
> > > > And what does the reply to 'qscreen' mean?
> > > > 
> > > > I'm guessing there is some Windows information page that has all
> > > > these definitions, and I hope someone can point me to it.
> > > > 
> > > > Henry Rich 
> > > > 
> > > > > -----Original Message-----
> > > > > From: programming-bounces@xxxxxxxxxxxxx 
> > > > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf 
> Of Henry Rich
> > > > > Sent: Sunday, December 31, 2006 2:37 PM
> > > > > To: 'Programming forum'
> > > > > Subject: RE: [Jprogramming] Form design for multiple 
> desktop sizes
> > > > > 
> > > > >  
> > > > > 
> > > > > > See screen metrics.
> > > > > 
> > > > > That sounds useful.  Where is it?
> > > > > 
> > > > > Henry Rich
> > > 
> > > 
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Tired of spam?  Yahoo! Mail has the best spam protection around 
> > > http://mail.yahoo.com 
> > > 
> ----------------------------------------------------------------------
> > > For information about J forums see 
> > > http://www.jsoftware.com/forums.htm
> > 
> > 
> ----------------------------------------------------------------------
> > For information about J forums see 
> http://www.jsoftware.com/forums.htm
> > 
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
> ----------------------------------------------------------------------
> For information about J forums see 
> http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: Oleg Kobchenko <olegykj@xxxxxxxxx>

In-reply-to: <004501c72d1a$a02b7270$6501a8c0@DD342W61 >
References: <004501c72d1a$a02b7270$6501a8c0@DD342W61 >

Esp. that you are comparing output of
qscreen with definition of qm.

--- Henry Rich <HenryHRich@xxxxxxxxx> wrote:

> I think I'm supposed to understand this stuff, but
> I just don't know what the words mean...    in
> 
> C:\J\system\extras\help\user\wd_commands.htm
> 
> I see
> 
> qm ; return system metrics:
>         screen width, screen height,
>         x logical unit, y logical unit,
>         cxborder, cyborder,
>         cxfixedframe, cyfixedframe,
>         cxframe, cyframe,
>         cycaption, cymenu,
>         desktop x, desktop y,
>         desktop w, desktop h
>         More elements may be added.
> 
> 
> I just don't know what this refers to.  For example,
> 'x logical unit'.  I don't know if this is describing
> something in logical units (in which case my question
> ins, What's a logical unit?) or telling me what a logical
> unit is (in which case I have the additional question,
> What is it that is being measured in logical units?).
> 
> On my system I get
> 
>    wd 'qscreen'
> 362 290 1280 1024 96 96 32 1 _1 36 36 51 
> 
> 12 values, which doesn't match the 16 things listed above.
> 
> I have similar basic questions about wd 'qscreen'.
> 
> 
> I just have the feeling this stuff must be described somewhere.
> 
> Henry Rich
>  
> 
> > -----Original Message-----
> > From: programming-bounces@xxxxxxxxxxxxx 
> > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Oleg Kobchenko
> > Sent: Sunday, December 31, 2006 3:25 PM
> > To: Programming forum
> > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > 
> > Yes to both, it's explained in WD Commands page.
> > 
> > The things they relate to is the basic
> > UI stuff, like the size of the window title bar etc.
> > 
> > --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> > 
> > > Wait... is it the reply to 'qm'?  What do all those things mean?
> > > And what does the reply to 'qscreen' mean?
> > > 
> > > I'm guessing there is some Windows information page that has all
> > > these definitions, and I hope someone can point me to it.
> > > 
> > > Henry Rich 
> > > 
> > > > -----Original Message-----
> > > > From: programming-bounces@xxxxxxxxxxxxx 
> > > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Henry Rich
> > > > Sent: Sunday, December 31, 2006 2:37 PM
> > > > To: 'Programming forum'
> > > > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > > > 
> > > >  
> > > > 
> > > > > See screen metrics.
> > > > 
> > > > That sounds useful.  Where is it?
> > > > 
> > > > Henry Rich


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: Oleg Kobchenko <olegykj@xxxxxxxxx>

In-reply-to: <004501c72d1a$a02b7270$6501a8c0@DD342W61 >
References: <004501c72d1a$a02b7270$6501a8c0@DD342W61 >

You don't need to know.

   wd'qm'
1280 1024 8 16 1 1 3 3 4 4 22 20 0 0 1280 994

   wd'pmovex 0 0 ',": _2{.". wd'qm'

will resize it to full desktop, which is
smaller than screen.

I get it not from knowledge, but from observation.


--- Henry Rich <HenryHRich@xxxxxxxxx> wrote:

> I think I'm supposed to understand this stuff, but
> I just don't know what the words mean...    in
> 
> C:\J\system\extras\help\user\wd_commands.htm
> 
> I see
> 
> qm ; return system metrics:
>         screen width, screen height,
>         x logical unit, y logical unit,
>         cxborder, cyborder,
>         cxfixedframe, cyfixedframe,
>         cxframe, cyframe,
>         cycaption, cymenu,
>         desktop x, desktop y,
>         desktop w, desktop h
>         More elements may be added.
> 
> 
> I just don't know what this refers to.  For example,
> 'x logical unit'.  I don't know if this is describing
> something in logical units (in which case my question
> ins, What's a logical unit?) or telling me what a logical
> unit is (in which case I have the additional question,
> What is it that is being measured in logical units?).
> 
> On my system I get
> 
>    wd 'qscreen'
> 362 290 1280 1024 96 96 32 1 _1 36 36 51 
> 
> 12 values, which doesn't match the 16 things listed above.
> 
> I have similar basic questions about wd 'qscreen'.
> 
> 
> I just have the feeling this stuff must be described somewhere.
> 
> Henry Rich
>  
> 
> > -----Original Message-----
> > From: programming-bounces@xxxxxxxxxxxxx 
> > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Oleg Kobchenko
> > Sent: Sunday, December 31, 2006 3:25 PM
> > To: Programming forum
> > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > 
> > Yes to both, it's explained in WD Commands page.
> > 
> > The things they relate to is the basic
> > UI stuff, like the size of the window title bar etc.
> > 
> > --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> > 
> > > Wait... is it the reply to 'qm'?  What do all those things mean?
> > > And what does the reply to 'qscreen' mean?
> > > 
> > > I'm guessing there is some Windows information page that has all
> > > these definitions, and I hope someone can point me to it.
> > > 
> > > Henry Rich 
> > > 
> > > > -----Original Message-----
> > > > From: programming-bounces@xxxxxxxxxxxxx 
> > > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Henry Rich
> > > > Sent: Sunday, December 31, 2006 2:37 PM
> > > > To: 'Programming forum'
> > > > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > > > 
> > > >  
> > > > 
> > > > > See screen metrics.
> > > > 
> > > > That sounds useful.  Where is it?
> > > > 
> > > > Henry Rich
> > 
> > 
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam protection around 
> > http://mail.yahoo.com 
> > ----------------------------------------------------------------------
> > For information about J forums see 
> > http://www.jsoftware.com/forums.htm
> 
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <781278.47825.qm@xxxxxxxxxxxxxxxxxxxxxxx>
References: <781278.47825.qm@xxxxxxxxxxxxxxxxxxxxxxx>

I think I'm supposed to understand this stuff, but
I just don't know what the words mean...    in

C:\J\system\extras\help\user\wd_commands.htm

I see

qm ; return system metrics:
        screen width, screen height,
        x logical unit, y logical unit,
        cxborder, cyborder,
        cxfixedframe, cyfixedframe,
        cxframe, cyframe,
        cycaption, cymenu,
        desktop x, desktop y,
        desktop w, desktop h
        More elements may be added.


I just don't know what this refers to.  For example,
'x logical unit'.  I don't know if this is describing
something in logical units (in which case my question
ins, What's a logical unit?) or telling me what a logical
unit is (in which case I have the additional question,
What is it that is being measured in logical units?).

On my system I get

   wd 'qscreen'
362 290 1280 1024 96 96 32 1 _1 36 36 51 

12 values, which doesn't match the 16 things listed above.

I have similar basic questions about wd 'qscreen'.


I just have the feeling this stuff must be described somewhere.

Henry Rich
 

> -----Original Message-----
> From: programming-bounces@xxxxxxxxxxxxx 
> [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Oleg Kobchenko
> Sent: Sunday, December 31, 2006 3:25 PM
> To: Programming forum
> Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> 
> Yes to both, it's explained in WD Commands page.
> 
> The things they relate to is the basic
> UI stuff, like the size of the window title bar etc.
> 
> --- Henry Rich <HenryHRich@xxxxxxxxx> wrote:
> 
> > Wait... is it the reply to 'qm'?  What do all those things mean?
> > And what does the reply to 'qscreen' mean?
> > 
> > I'm guessing there is some Windows information page that has all
> > these definitions, and I hope someone can point me to it.
> > 
> > Henry Rich 
> > 
> > > -----Original Message-----
> > > From: programming-bounces@xxxxxxxxxxxxx 
> > > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Henry Rich
> > > Sent: Sunday, December 31, 2006 2:37 PM
> > > To: 'Programming forum'
> > > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > > 
> > >  
> > > 
> > > > See screen metrics.
> > > 
> > > That sounds useful.  Where is it?
> > > 
> > > Henry Rich
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
> ----------------------------------------------------------------------
> For information about J forums see 
> http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: Oleg Kobchenko <olegykj@xxxxxxxxx>

In-reply-to: <004301c72d13$70a182d0$6501a8c0@DD342W61 >
References: <004301c72d13$70a182d0$6501a8c0@DD342W61 >

Yes to both, it's explained in WD Commands page.

The things they relate to is the basic
UI stuff, like the size of the window title bar etc.

--- Henry Rich <HenryHRich@xxxxxxxxx> wrote:

> Wait... is it the reply to 'qm'?  What do all those things mean?
> And what does the reply to 'qscreen' mean?
> 
> I'm guessing there is some Windows information page that has all
> these definitions, and I hope someone can point me to it.
> 
> Henry Rich 
> 
> > -----Original Message-----
> > From: programming-bounces@xxxxxxxxxxxxx 
> > [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Henry Rich
> > Sent: Sunday, December 31, 2006 2:37 PM
> > To: 'Programming forum'
> > Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> > 
> >  
> > 
> > > See screen metrics.
> > 
> > That sounds useful.  Where is it?
> > 
> > Henry Rich


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <004201c72d13$19a2d9c0$6501a8c0@DD342W61 >
References: <004201c72d13$19a2d9c0$6501a8c0@DD342W61 >

Wait... is it the reply to 'qm'?  What do all those things mean?
And what does the reply to 'qscreen' mean?

I'm guessing there is some Windows information page that has all
these definitions, and I hope someone can point me to it.

Henry Rich 

> -----Original Message-----
> From: programming-bounces@xxxxxxxxxxxxx 
> [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Henry Rich
> Sent: Sunday, December 31, 2006 2:37 PM
> To: 'Programming forum'
> Subject: RE: [Jprogramming] Form design for multiple desktop sizes
> 
>  
> 
> > See screen metrics.
> 
> That sounds useful.  Where is it?
> 
> Henry Rich
> 
> ----------------------------------------------------------------------
> For information about J forums see 
> http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <51406.80911.qm@xxxxxxxxxxxxxxxxxxxxxxx>
References: <51406.80911.qm@xxxxxxxxxxxxxxxxxxxxxxx>

 

> See screen metrics.

That sounds useful.  Where is it?

Henry Rich

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: Oleg Kobchenko <olegykj@xxxxxxxxx>

In-reply-to: <003f01c72d09$0fdff6c0$6501a8c0@DD342W61 >
References: <003f01c72d09$0fdff6c0$6501a8c0@DD342W61 >

--- Henry Rich <HenryHRich@xxxxxxxxx> wrote:

> For the first time I have to deploy an application across
> a variety of machines with a variety of desktop sizes.
> I wonder how I should design my forms to make this work.
> 
> I would like certain forms to almost fill the screen regardless
> of desktop size.  But they don't: on smaller desktops they are
> too big, on big ones they are too small.
> 
> My questions:
> 
> 0.  What are the units of the numbers the Form Editor produces.
>   They don't seem to be pixels, but they seem to be related to pixels.

It might be 2 pixel units.

> 1.  Is there a way I am supposed to design my forms to make the form
>   sizes a fixed proportion of the screen size?

Design a form to fit minimal screen.
Then after form is shown, query screen size
and programmatically resize the form.

> 2.  I suppose it would be easy enough to scale all the numbers in
>   a form using something like
>   <.@(scalefactor&*)&.".^:([: *./ '0123456789 ' e.~ ])&.>&.;: FORM
>   if I could figure out the scale factor.  wd 'qscreen' seems to
>   be applicable.  What is a logical inch as used in qscreen?
>
> Henry Rich
> 

Take into account both form and screen adornments (task bar, etc).
See screen metrics.


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] Form editor doesn't honor ss_right in static control

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <458F051A.8080602@xxxxxxxxxxxxx>
References: <458F051A.8080602@xxxxxxxxxxxxx>

When I assign the ss_right attribute to a static, the Form Editor
doesn't right-align the text.  When the form is ultimately run,
it is correctly aligned, but not during design.

Example below,

Henry Rich




RIMSIM=: 0 : 0
pc rimsim;
xywh 4 17 35 10;cc labpc static ss_right;cn "PC";
xywh 40 16 32 11;cc fmpc edit;
xywh 4 37 35 10;cc labr0 static ss_right;cn "PC";
xywh 40 36 32 11;cc fmr0 edit;
pas 6 6;pcenter;
rem form end;
)

rimsim_run=: 3 : 0
wd RIMSIM
NB. initialize form here
wd 'pshow;'
)

rimsim_close=: 3 : 0
wd'pclose'
)

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] Inverse for m&".

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <458F051A.8080602@xxxxxxxxxxxxx>
References: <458F051A.8080602@xxxxxxxxxxxxx>

   0&". b. _1
|domain error

Wouldn't it be a good idea for ": to be the inverse of m&".  ?

Henry Rich

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] Form design for multiple desktop sizes

Date: December 31, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <458F051A.8080602@xxxxxxxxxxxxx>
References: <458F051A.8080602@xxxxxxxxxxxxx>

For the first time I have to deploy an application across
a variety of machines with a variety of desktop sizes.
I wonder how I should design my forms to make this work.

I would like certain forms to almost fill the screen regardless
of desktop size.  But they don't: on smaller desktops they are
too big, on big ones they are too small.

My questions:

0.  What are the units of the numbers the Form Editor produces.
  They don't seem to be pixels, but they seem to be related to pixels.

1.  Is there a way I am supposed to design my forms to make the form
  sizes a fixed proportion of the screen size?

2.  I suppose it would be easy enough to scale all the numbers in
  a form using something like
  <.@(scalefactor&*)&.".^:([: *./ '0123456789 ' e.~ ])&.>&.;: FORM
  if I could figure out the scale factor.  wd 'qscreen' seems to
  be applicable.  What is a logical inch as used in qscreen?

Henry Rich

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Boxing/unboxing

Date: December 31, 2006
From: Roger Hui <rhui000@xxxxxxx>

See also http://www.jsoftware.com/jwiki/Essays/Under



----- Original Message -----
From: June Kim <juneaftn@xxxxxxxxx>
Date: Sunday, December 31, 2006 4:17 am
Subject: Re: [Jprogramming] Boxing/unboxing

> Cute examples of under:
> 
>   dfh=. 16&#. @ ('0123456789ABCDEF'&i.)    NB. decimal from hex
>   hex=. &. dfh
> 
>   'FEED' + hex 'B'
> FEF8
> 
> http://www.jsoftware.com/papers/elegant.htm
> 
> 
>      ((i.4 4)+])&.(a.&i.) 'A'
> ABCD
> EFGH
> IJKL
> MNOP
 

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Boxing/unboxing

Date: December 31, 2006
From: "June Kim" <juneaftn@xxxxxxxxx>

In-reply-to: <6f80c1520612310411g36a451e1u11f5bf24c6c5db69@xxxxxxxxxxxxxx>
References: <BHEAJFOCJBDNKGLGIFAAAEONCAAA.asandstrom@xxxxxxxxxxxxx> <6f80c1520612310411g36a451e1u11f5bf24c6c5db69@xxxxxxxxxxxxxx>

Cute examples of under:

  dfh=. 16&#. @ ('0123456789ABCDEF'&i.)    NB. decimal from hex
  hex=. &. dfh

  'FEED' + hex 'B'
FEF8

http://www.jsoftware.com/papers/elegant.htm


     ((i.4 4)+])&.(a.&i.) 'A'
ABCD
EFGH
IJKL
MNOP


2006/12/31, June Kim <juneaftn@xxxxxxxxx>:

What you need is "under".

     s=. 'tratatrittetratas'
     i=.2 3 4;7 8 9 10;13 14 15 16
     {&s each i
+---+----+----+
|ata|itte|atas|
+---+----+----+

each is pre-defined as &.>

2006/12/31, Arved Sandstrom <asandstrom@xxxxxxxxxxxxx>:
> Best of the New Year to all,
>
> I'm having some problems getting a handle on boxing and unboxing. I am still
> in the novice stages with J - I don't have much time to devote to the
> language, unfortunately, and sometimes go a few weeks at a time without
> looking at it. The holidays gave me an opportunity to spend some hours to
> tackle J again.
>
> I usually take a set of basic problems to tackle when learning a new
> language, and one of them is splitting a string on another string, no
> regular expressions involved. Example:
>
> 'tr' split 'tratatrittetratas'
>
> should return
>
> 'ata' 'itte' 'atas'
>
> I have the J code to the point that for the above example I generate
>
>  --------------------------------
> | 2 3 4 | 7 8 9 10 | 13 14 15 16 |
>  --------------------------------
>
> and any of the three boxes, when unboxed and applied to the string with {,
> gives me the substring I need.
>
> What's causing me problems is fills. I have not happened on a combination of
> < and > that would allow me to return
>
>  ---------------------------
> | a t a | i t t e | a t a s |
>  ---------------------------
>
> Instead I might get
>
> atat
> itte
> atas
>
> which is frustrating. Any ideas?
>
> AHS
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Boxing/unboxing

Date: December 31, 2006
From: "June Kim" <juneaftn@xxxxxxxxx>

In-reply-to: <BHEAJFOCJBDNKGLGIFAAAEONCAAA.asandstrom@xxxxxxxxxxxxx>
References: <BHEAJFOCJBDNKGLGIFAAAEONCAAA.asandstrom@xxxxxxxxxxxxx>

What you need is "under".

     s=. 'tratatrittetratas'
     i=.2 3 4;7 8 9 10;13 14 15 16
     {&s each i
+---+----+----+
|ata|itte|atas|
+---+----+----+

each is pre-defined as &.>

2006/12/31, Arved Sandstrom <asandstrom@xxxxxxxxxxxxx>:

Best of the New Year to all,

I'm having some problems getting a handle on boxing and unboxing. I am still
in the novice stages with J - I don't have much time to devote to the
language, unfortunately, and sometimes go a few weeks at a time without
looking at it. The holidays gave me an opportunity to spend some hours to
tackle J again.

I usually take a set of basic problems to tackle when learning a new
language, and one of them is splitting a string on another string, no
regular expressions involved. Example:

'tr' split 'tratatrittetratas'

should return

'ata' 'itte' 'atas'

I have the J code to the point that for the above example I generate

 --------------------------------
| 2 3 4 | 7 8 9 10 | 13 14 15 16 |
 --------------------------------

and any of the three boxes, when unboxed and applied to the string with {,
gives me the substring I need.

What's causing me problems is fills. I have not happened on a combination of
< and > that would allow me to return

 ---------------------------
| a t a | i t t e | a t a s |
 ---------------------------

Instead I might get

atat
itte
atas

which is frustrating. Any ideas?

AHS
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] Boxing/unboxing

Date: December 31, 2006
From: Arved Sandstrom <asandstrom@xxxxxxxxxxxxx>

Best of the New Year to all,

I'm having some problems getting a handle on boxing and unboxing. I am still
in the novice stages with J - I don't have much time to devote to the
language, unfortunately, and sometimes go a few weeks at a time without
looking at it. The holidays gave me an opportunity to spend some hours to
tackle J again.

I usually take a set of basic problems to tackle when learning a new
language, and one of them is splitting a string on another string, no
regular expressions involved. Example:

'tr' split 'tratatrittetratas'

should return

'ata' 'itte' 'atas'

I have the J code to the point that for the above example I generate

 --------------------------------
| 2 3 4 | 7 8 9 10 | 13 14 15 16 |
 --------------------------------

and any of the three boxes, when unboxed and applied to the string with {,
gives me the substring I need.

What's causing me problems is fills. I have not happened on a combination of
< and > that would allow me to return

 ---------------------------
| a t a | i t t e | a t a s |
 ---------------------------

Instead I might get

atat
itte
atas

which is frustrating. Any ideas?

AHS
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] problem with regex

Date: December 31, 2006
From: "david alis" <david.alis@xxxxxxxxx>

The following crashes my machine (windows XP ) running J
j601/2006-11-17/17:05


require 'regex'
pat =: '.*:[:alpha:]+:[:alpha:]+:'
data =:
'ARR++A:N:8:N:A:US:A:6:19752004:702:0:A+0:A+0:A+0:A+0:A+0:A+0:A+0:A+0:A+0:A+0:A+174233:A+179404:A+119607:A+130989:A+151837:A+145934:A+155365:A+182601:A+294121:A+379634:A+399257:A+461510:A+529953:A+621692:A+829887:A+949264:A+994197:A+1129040:A+1468940:A'

NB. 251 -: #data (no white space in data)

pat rxmatch data


Regards
David
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] how to fill the sparse matrix

Date: December 31, 2006
From: "Kairit Sirts" <karambula@xxxxxx>

In-reply-to: <20061230183747.02DC074C641@xxxxxxxxxxxxxxxxxxx>
References: <20061230183747.02DC074C641@xxxxxxxxxxxxxxxxxxx>

The problem is that I cant make a1 and a2 sparse, because they consist of
boxed strings. 

Kairit

> 
> > How should I calculate [the following] for my sparse matrix?
> > Bigarr =: a1 =/ a2.
> 
> Well, I would've advised you to first make  a1  and  a2  sparse
> themselves, i.e.:
> 
>          a1 =/&:$. a2.
> 
> But, when I tried some experiments:
> 
>          a1 =. i. 4000
>          a2 =. 60e3 $ 78
> 
>          $ a1 =/ a2
>       4000 60000
>          sa1 =. $. a1
>          sa2 =. $. a2
> 
>          $ sa1 =/ sa2
>       |non-unique sparse elements: scriptd
>       |   $sa1    =/sa2
>       |[-7] c:\app\dev\lang\j\601c\temp\13.ijs
> 
> Which looks like a bug (to me), and is reducible to :
> 
>          =/~ $. , 1
>       |non-unique sparse elements
>       |       =/~$.,1
> -Dan
> 
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] text replacement without RE

Date: December 31, 2006
From: "June Kim" <juneaftn@xxxxxxxxx>

In-reply-to: <6d07b26cf082.6cf0826d07b2@xxxxxxx>
References: <6d07b26cf082.6cf0826d07b2@xxxxxxx>

2006/12/31, Roger Hui <rhui000@xxxxxxx>:

enc=: 4 : 0
 m=. (-2&|) +/ b=. x E. y
 j=. (i.m)+(m$0,#x)+m$I. b
 (m$'{}') j} (-.(i.m+#y) e. j)#^:_1 y
)


Nice. I didn't think of using "spread" here. I'd done it my way with
cut and connect the fragments, but yours is, of course, much neater.

[snip]


In such cases, the E. in enc needs to be replaced
by nos from the "Non-Overlapping Substrings" essay
http://www.jsoftware.com/jwiki/Essays/Non-Overlapping_Substrings


The use of transitive closure is truly enlightening. (I have your
"Some Uses of { and }" paper, too)

I take your code as a new year's gift. Thank you, Roger!

[snip]


----- Original Message -----
From: June Kim <juneaftn@xxxxxxxxx>
Date: Saturday, December 30, 2006 5:25 pm
Subject: [Jprogramming] text replacement without RE

> Could you come up with a verb enc, which encloses each substring that
> starts and ends with x in y without using RE? The substrings are
> discovered from the left.
>
> Example:
>
>   'foo' enc 'x xfoo x x xxfoo gogo foofoo barfoo'
> 'x x{foo x x xxfoo} gogo {foofoo} barfoo'
>
> Happy new year!


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] text replacement without RE

Date: December 31, 2006
From: Roger Hui <rhui000@xxxxxxx>

enc=: 4 : 0
 m=. (-2&|) +/ b=. x E. y
 j=. (i.m)+(m$0,#x)+m$I. b
 (m$'{}') j} (-.(i.m+#y) e. j)#^:_1 y
)

x=: 'foo'
y=: 'x xfoo x x xxfoo gogo foofoo barfoo'

   x enc y
 x{foo x x xxfoo} gogo {foofoo} barfoo

However, if the substring occurrences overlap, enc
gives wrong answers:

   'abab' enc 'ababab inside ababab outside'
{ababab} inside {ababab} outside

In such cases, the E. in enc needs to be replaced 
by nos from the "Non-Overlapping Substrings" essay
http://www.jsoftware.com/jwiki/Essays/Non-Overlapping_Substrings
Thus:

nos=: 4 : 0
 s=. x I.@E. y
 i=. s I. s+#x
 (i.#y) e. (s,_1) {~ {&(i,_1)^:a: 0
)

enc1=: 4 : 0
 m=. (-2&|) +/ b=. x nos y
 j=. (i.m)+(m$0,#x)+m$I. b
 (m$'{}') j} (-.(i.m+#y) e. j)#^:_1 y
)

   'abab' enc1 'ababab inside ababab outside'
{ababab inside abab}ab outside
   x enc1 y
x x{foo x x xxfoo} gogo {foofoo} barfoo



----- Original Message -----
From: June Kim <juneaftn@xxxxxxxxx>
Date: Saturday, December 30, 2006 5:25 pm
Subject: [Jprogramming] text replacement without RE

> Could you come up with a verb enc, which encloses each substring that
> starts and ends with x in y without using RE? The substrings are
> discovered from the left.
> 
> Example:
> 
>   'foo' enc 'x xfoo x x xxfoo gogo foofoo barfoo'
> 'x x{foo x x xxfoo} gogo {foofoo} barfoo'
> 
> Happy new year!


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] text replacement without RE

Date: December 31, 2006
From: "June Kim" <juneaftn@xxxxxxxxx>

Could you come up with a verb enc, which encloses each substring that
starts and ends with x in y without using RE? The substrings are
discovered from the left.

Example:

  'foo' enc 'x xfoo x x xxfoo gogo foofoo barfoo'
'x x{foo x x xxfoo} gogo {foofoo} barfoo'

Happy new year!
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

December 30, 2006

RE: [Jprogramming] Matrix Multiplication

Date: December 30, 2006
From: "Henry Rich" <HenryHRich@xxxxxxxxx>

In-reply-to: <000701c72c5e$b0e9aa80$0b02a8c0@PENTIUM424 >
References: <000701c72c5e$b0e9aa80$0b02a8c0@PENTIUM424 >

+/ . * "2

Henry Rich 

> -----Original Message-----
> From: programming-bounces@xxxxxxxxxxxxx 
> [mailto:programming-bounces@xxxxxxxxxxxxx] On Behalf Of Carl F. Buhrer
> Sent: Saturday, December 30, 2006 5:06 PM
> To: programming@xxxxxxxxxxxxx
> Subject: [Jprogramming] Matrix Multiplication
> 
> If $ m1 equals 7 3 4 and $ m2 equals 7 4 5 where m1 and m2 
> each represent 7 
> matricies,  how do I construct
> mm , a matrix multiplication verb that forms m3 representing 7 matrix 
> products with  $ m3 equal to 7 3 5   ?
> I can select a pair at a time and use +/ . * within a do. 
> control loop,  but 
> wonder whether there is a simpler way.
>                                                         C. F. Buhrer 
> ----------------------------------------------------------------------
> For information about J forums see 
> http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] Matrix Multiplication

Date: December 30, 2006
From: "Carl F. Buhrer" <cfbuhrer@xxxxxxx>

If $ m1 equals 7 3 4 and $ m2 equals 7 4 5 where m1 and m2 each represent 7

matricies, how do I construct

mm , a matrix multiplication verb that forms m3 representing 7 matrix

products with $ m3 equal to 7 3 5 ?

I can select a pair at a time and use +/ . * within a do. control loop, but

wonder whether there is a simpler way.

C. F. Buhrer

----------------------------------------------------------------------

For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: Eugene McDonnell <eemcd@xxxxxxx>

In-reply-to: <4595FA86.5060503@xxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx> <20061230003837.70BF85BC115@xxxxxxxxxxxxxxx> <4595FA86.5060503@xxxxxxxxxxxxx>


On Dec 29, 2006, at 9:35 PM, Chris Burke wrote:

Nevertheless, it is important that the date routines be reliable, and
the suggestion that they may not be is a concern. Are there standard
tests that should be run?

I use the sum of the number of days in any 400 consecutive years in

the Gregorian calendar. It should amount to 146097, which, divided

by 400 yields 365.2425 for the average length of the Gregorian year.

The current estimate of astronomers for year length is 365.242199See

the essay "How Long is a Year" in APL Quote Quad 7.4, Winter 1977.


Eugene
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] how to fill the sparse matrix

Date: December 30, 2006
From: "Dan Bron" <jf@xxxxxxxxx>

In-reply-to: <20061230175429.1CD5D32C433@xxxxxxxxxxxxxx>
References: <20061230175429.1CD5D32C433@xxxxxxxxxxxxxx>

> How should I calculate [the following] for my sparse matrix?
> Bigarr =: a1 =/ a2.

Well, I would've advised you to first make  a1  and  a2  sparse themselves, 
i.e.:

           a1 =/&:$. a2.

But, when I tried some experiments:

           a1 =. i. 4000
           a2 =. 60e3 $ 78

           $ a1 =/ a2
        4000 60000
           sa1 =. $. a1
           sa2 =. $. a2
      
           $ sa1 =/ sa2
        |non-unique sparse elements: scriptd
        |   $sa1    =/sa2
        |[-7] c:\app\dev\lang\j\601c\temp\13.ijs
   
Which looks like a bug (to me), and is reducible to :
           
           =/~ $. , 1
        |non-unique sparse elements
        |       =/~$.,1
-Dan

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

RE: [Jprogramming] regex questions

Date: December 30, 2006
From: "Dan Bron" <jf@xxxxxxxxx>

In-reply-to: <200612291750.kBTHoMQY022458@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
References: <200612291750.kBTHoMQY022458@xxxxxxxxxxxxxxxxxxxxxxxxxxx>

Phillip wrote:
>  I want to replace the unadorned form ('viton') with the initial'd form 
> ('viton,p'). 

If you're simply replacing (<'viton') with (<'viton,p'), skip the regexes, and 
just use:
  
      here  =. <'viton'
      there =. <'viton,p'
      array =: 500 (] {~ (?.@$ #)) here,there

      ~. array
        +-----+-------+
        |viton|viton,p|
        +-----+-------+

      fixed =: here&=`(,:&there)} array

      ~. fixed
        +-------+
        |viton,p|
        +-------+

-Dan

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

[Jprogramming] how to fill the sparse matrix

Date: December 30, 2006
From: "Kairit Sirts" <karambula@xxxxxx>

Hi!

I have to make a 2-dimensional matrix, which is very big (1000 * app. 2 mio)
and consists mainly from 0 and 1 in some places. So I understand that most
reasonable is to make it as sparse.

However, I don't understand how should I calculate the values of such a
sparse matrix. Basically I want to calculate it in the following way:
I have 2 arrays: a1 and a2 and this big array is calculated as:

Bigarr =: a1 =/ a2.

How should I calculate it for my sparse matrix?

Kairit


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: Anssi Sepp�l� <anssi.seppala@xxxxxxxxx>

In-reply-to: <4595FA86.5060503@xxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx> <20061230003837.70BF85BC115@xxxxxxxxxxxxxxx> <4595FA86.5060503@xxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx> <20061230003837.70BF85BC115@xxxxxxxxxxxxxxx> <4595FA86.5060503@xxxxxxxxxxxxx>

After working 20 years with dates and calendar data I agree that J's

date functions are reliable and I rely on them calculating

differences between dates. Holidays and week numbers are different case.

At 07:35 30.12.2006, you wrote:

I don't think anyone has yet found a problem with J's date handling.


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: bill lam <bbill.lam@xxxxxxxxx>

In-reply-to: <4595FA86.5060503@xxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx> <20061230003837.70BF85BC115@xxxxxxxxxxxxxxx> <4595FA86.5060503@xxxxxxxxxxxxx>

Chris Burke wrote:
> I don't think anyone has yet found a problem with J's date handling.
> This thread was initiated by someone who thought there was an error, but
> quickly admitted his mistake.
> 
> Nevertheless, it is important that the date routines be reliable, and
> the suggestion that they may not be is a concern. Are there standard
> tests that should be run? Obvious ones like the following seem to work
> correctly:

I think J's date routines is OK. A very minor issue is that verb "todate" does
not <. its argument. eg,
   todayno 2006 12 31
75604
   todate 75604.1
2006 12 31
   todate 75604.5
2007 1 1

Day should be floored rather than rounded because the day 2006 12 31 should
include midnight 00:00:00 upto 23:59:59 before next midnight. Not a bug but only
that users should remember to apply <. before passing argument to "todate"

-- 
regards,
bill
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: Chris Burke <cburke@xxxxxxxxxxxxx>

In-reply-to: <20061230003837.70BF85BC115@xxxxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx> <20061230003837.70BF85BC115@xxxxxxxxxxxxxxx>

Anssi Seppälä wrote:
> The leap year is one case, but there is more: eastern, ISO week numbers
> etc. that are often needed. My experience goes to the 1980's when I
> found an APL function DATUM. It did great job many years until it did
> wrong somewhere after 1990.  The conclusion is that no programmer has
> the time to carefully check the algorithm over many future years.
> Writers of Calendrical Tabulations have done the job seriously and
> professionally and therefore I trust on them. When I use calendrical
> information in my applications I allways check it with published calendars.
> 
> Anssi
> 
> At 02:21 30.12.2006, you wrote:
>> Are there any particular examples how this book is
>> more precise than following the mentioned formula,
>> or using the dates script.
>>
>> --- Anssi Seppälä <anssi.seppala@xxxxxxxxx> wrote:
>>
>> > My experience is that the APL/J calendrical algorithms are allways
>> > unreliable. The testing is allways insufficient. My choice was to buy
>> > Reingold&Dershowiz: Calendrical Tabulations 1900 - 2200. There I find
>> > reliable data and put that into the applications.

I don't think anyone has yet found a problem with J's date handling.
This thread was initiated by someone who thought there was an error, but
quickly admitted his mistake.

Nevertheless, it is important that the date routines be reliable, and
the suggestion that they may not be is a concern. Are there standard
tests that should be run? Obvious ones like the following seem to work
correctly:

   load 'dates'
   a=. (i.1e5) + todayno 1900 1 1
   b=. todate a
   a -: todayno b
1
   {:b
2173 10 15
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: "PackRat" <packrat@xxxxxxxx>

In-reply-to: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx>
References: <4031fef0612290527n3ed47f62t4e64dcc07a87adaf@xxxxxxxxxxxxxx>, <BAC49A00-2EEB-47A4-9B35-185C8BCA2A3D@xxxxxxx>, <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx>

Anssi Seppälä wrote:
> My experience is that the APL/J calendrical algorithms are allways
> unreliable. The testing is allways insufficient. My choice was to buy
> Reingold&Dershowiz: Calendrical Tabulations 1900 - 2200. There I find
> reliable data and put that into the applications.

FWIW with regard to this discussion:

I believe that the better book is the 2nd edition of the same authors'
"Calendrical Calculations" because it gives the formulas and algorithms
that one would use in computer programming.  The story of trying to
"nail jelly to a tree" for some of the world's calendar systems is a
quite interesting narrative in itself because of all the exceptions and
sometimes near impossibility of translating them to computer code (Lisp
and Java).  You can see the Amazon citation at:
<http://www.amazon.com/Calendrical-Calculations-Millennium-Edward-
Reingold/dp/0521777526>.  This book should be in every programmer's
"top 100" (maybe even "top 25").

The only other book worth its weight in gold for information on time-
related phenomena (such as calendars and other celestial timings) is
the "Explanatory Supplement to the Astronomical Almanac".  For obvious
reasons, astronomers are extremely concerned about accuracy related to
time.  The Amazon citation is at: <http://www.amazon.com/Explanatory-
Supplement-Astronomical-Almanac-Seidelmann/dp/1891389459>.

Harvey

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Leap Year

Date: December 30, 2006
From: bill lam <bbill.lam@xxxxxxxxx>

In-reply-to: <00f401c72b73$a7e25590$f7a700b0$@it >
References: <4031fef0612290527n3ed47f62t4e64dcc07a87adaf@xxxxxxxxxxxxxx> <45951CD0.2050107@xxxxxxxxx> <45953C32.8000404@xxxxxxxxxxxx> <4031fef0612290848l315b7905g90a70241872eb5f3@xxxxxxxxxxxxxx> <00f001c72b70$11603e00$3420ba00$@it > <6dd16b0a0612290957t4a44d193p321221975bb3ced1@xxxxxxxxxxxxxx> <00f401c72b73$a7e25590$f7a700b0$@it >

Stefano Lanzavecchia wrote:

As far as I can tell the flag is saved per-workbook. And, I guess, it's a
change that required a new file format, incompatible with the previous
versions of Excel.
http://msdn2.microsoft.com/en-us/library/aa204992(office.10).aspx
http://support.microsoft.com/kb/214330/ The knowledge base article seems to
hint that the possibility has existed at least since Excel 97.

AFAIK date1904 exists since biff2 (excel 2) at least 15 years ago I guess.

Apparently there is no way to enable it from excel program. However you can

enable date1904 when writing with tara. Eg.


  bi=. '' conew 'biffbook'      NB. create an workbook
  date1904__bi=: 1              NB. enable 1904

The next version of excel should use xml file format but I don't know if this

bug will be retained for compatibility.


--
regards,
bill
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: Anssi Sepp�l� <anssi.seppala@xxxxxxxxx>

In-reply-to: <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx> <460605.4423.qm@xxxxxxxxxxxxxxxxxxxxxxx>

The leap year is one case, but there is more: eastern, ISO week

numbers etc. that are often needed. My experience goes to the 1980's

when I found an APL function DATUM. It did great job many years until

it did wrong somewhere after 1990. The conclusion is that no

programmer has the time to carefully check the algorithm over many

future years. Writers of Calendrical Tabulations have done the job

seriously and professionally and therefore I trust on them. When I

use calendrical information in my applications I allways check it

with published calendars.


Anssi

At 02:21 30.12.2006, you wrote:

Are there any particular examples how this book is
more precise than following the mentioned formula,
or using the dates script.

--- Anssi Seppälä <anssi.seppala@xxxxxxxxx> wrote:

> My experience is that the APL/J calendrical algorithms are allways
> unreliable. The testing is allways insufficient. My choice was to buy
> Reingold&Dershowiz: Calendrical Tabulations 1900 - 2200. There I find
> reliable data and put that into the applications.
> Anssi


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: Oleg Kobchenko <olegykj@xxxxxxxxx>

In-reply-to: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx>
References: <20061230000434.433EB5BC0EC@xxxxxxxxxxxxxxx>

Are there any particular examples how this book is
more precise than following the mentioned formula,
or using the dates script.

--- Anssi Seppälä <anssi.seppala@xxxxxxxxx> wrote:

> My experience is that the APL/J calendrical algorithms are allways 
> unreliable. The testing is allways insufficient. My choice was to buy 
> Reingold&Dershowiz: Calendrical Tabulations 1900 - 2200. There I find 
> reliable data and put that into the applications.
> Anssi


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 30, 2006
From: Anssi Sepp�l� <anssi.seppala@xxxxxxxxx>

In-reply-to: <BAC49A00-2EEB-47A4-9B35-185C8BCA2A3D@xxxxxxx>
References: <4031fef0612290527n3ed47f62t4e64dcc07a87adaf@xxxxxxxxxxxxxx> <4031fef0612290548s4971ced1q1bd6399062dcbe62@xxxxxxxxxxxxxx> <6dd16b0a0612290908n1492bd97jb8faef51dec95d67@xxxxxxxxxxxxxx> <45958972.21146.B0CA977@xxxxxxxxxxxxxxxxxxxxxxxx> <BAC49A00-2EEB-47A4-9B35-185C8BCA2A3D@xxxxxxx>
References: <4031fef0612290527n3ed47f62t4e64dcc07a87adaf@xxxxxxxxxxxxxx> <4031fef0612290548s4971ced1q1bd6399062dcbe62@xxxxxxxxxxxxxx> <6dd16b0a0612290908n1492bd97jb8faef51dec95d67@xxxxxxxxxxxxxx> <45958972.21146.B0CA977@xxxxxxxxxxxxxxxxxxxxxxxx> <BAC49A00-2EEB-47A4-9B35-185C8BCA2A3D@xxxxxxx>

My experience is that the APL/J calendrical algorithms are allways

unreliable. The testing is allways insufficient. My choice was to buy

Reingold&Dershowiz: Calendrical Tabulations 1900 - 2200. There I find

reliable data and put that into the applications.

Anssi

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

December 29, 2006

Re: [Jprogramming] Re: Leap Year

Date: December 29, 2006
From: Eugene McDonnell <eemcd@xxxxxxx>

In-reply-to: <45958972.21146.B0CA977@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <4031fef0612290527n3ed47f62t4e64dcc07a87adaf@xxxxxxxxxxxxxx> <4031fef0612290548s4971ced1q1bd6399062dcbe62@xxxxxxxxxxxxxx> <6dd16b0a0612290908n1492bd97jb8faef51dec95d67@xxxxxxxxxxxxxx> <45958972.21146.B0CA977@xxxxxxxxxxxxxxxxxxxxxxxx>


On Dec 29, 2006, at 1:32 PM, Fred Bone wrote:

On 29 Dec 2006 at 12:08, Devon McCormick said:

Someone I know once calculated that there should be a 4000-yearcorrectionas well, to whit: there is an exception (flip parity) to the"divisible by
400" exception if the year is divisible by 4000.


Less than optimal.

The current setup is equivalent to a mean year of 365.2425 days. The
actual mean tropical year is 365.242199 days. The discrepancy is just
over 26 seconds a year, or 1 day every
   %365.2425-365.242199
3322.259136
years.

Omitting a leap year every 4000 years under-corrects, giving a meanyear

of
   365+-/%4 100 400 4000
365.24225
days, which is still too long by 1 day every
   %365.24225-365.242199
19607.84314
years: you need to omit another leap year at about this interval (say,
every 20000 years).

A better match IMHO is to omit once every 3200 years. This gives amean

year length of
   365+-/%4 100 400 3200
365.2421875
days, which is too short by one day in
   %365.2421875-~365.242199
86956.52153
years, which doesn't seem to admit a neat adjustment (though putting a
leap year back every 86400 years has a certain attraction).

However, since any change to the current calendar requires agreement
among at least most of the world's governments, it probably won't get
dealt with any time soon.

I wrote the essay "How Long is a Year" in APL Quote Quad 7.4, Winter

1977. It shows how to determine whether a year is common or leap,

assuming that the corrections occur in every 4th year, but not every

100th year unless it is also divisible by 400. It speculates about

extending this to include not every 3200th year unless it is the

86400th year. Such a correction would be one day off in a little over

thirteen and a half million years.

The article points out that 86400 is also the number of seconds in a

day -- which caused a Danish woman to have an intellectual orgasm. It

also led Jeffrey Shallit to write a joint paper with Erdos, which,

since I had written a joint paper with Jeff, gave me the Erdos number 2.

Phrase 14.B.m11 helps answer the question Is Year Y a Leap Year or

Common.




Eugene
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Re: [Jprogramming] Re: Leap Year

Date: December 29, 2006
From: Roger Hui <rhui000@xxxxxxx>

With the  noun verb verb  forks new in J6.01,
the expressions can be simplified from
   m11   =: 0: ~:/ .= 4 100 400"_ |/ ]
   m11_4k=: 0: ~:/ .= 4 100 400 4000"_ |/ ]
to
   m11   =: 0 ~:/ .= 4 100 400 |/ ]
   m11_4k=: 0 ~:/ .= 4 100 400 4000 |/ ]



----- Original Message -----
From: Devon McCormick <devonmcc@xxxxxxxxx>
Date: Friday, December 29, 2006 9:08 am
Subject: Re: [Jprogramming] Re: Leap Year

> Someone I know once calculated that there should be a 4000-year 
> correctionas well, to whit:
> there is an exception (flip parity) to the "divisible by 400" 
> exception if
> the year is divisible by 4000.
> 
> However, this assumes that the year calculation is purely 
> mathematical when
> in fact it is
> ultimately political.
> 
> The J expression "m11" is nicely succinct and easily extendible if 
> we wish
> to add this additional
> exception:
> 
>   m11=: 0: ~:/ .= 4 100 400"_ |/ ]
>   m11_4k=: 0: ~:/ .= 4 100 400 4000"_ |/ ]
> 
>   m11 1900 1901 1999 2000 2003 2004 3600 3999 4000 4004
> 0 0 0 1 0 1 1 0 1 1
>   m11_4k 1900 1901 1999 2000 2003 2004 3600 3999 4000 4004
> 0 0 0 1 0 1 1 0 0 1


----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm